CN108846823A - A kind of fusion method of terahertz image and visible images - Google Patents
A kind of fusion method of terahertz image and visible images Download PDFInfo
- Publication number
- CN108846823A CN108846823A CN201810654322.6A CN201810654322A CN108846823A CN 108846823 A CN108846823 A CN 108846823A CN 201810654322 A CN201810654322 A CN 201810654322A CN 108846823 A CN108846823 A CN 108846823A
- Authority
- CN
- China
- Prior art keywords
- image
- visible images
- terahertz
- terahertz image
- visible
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000007500 overflow downdraw method Methods 0.000 title claims abstract description 18
- 238000012937 correction Methods 0.000 claims abstract description 14
- 238000007781 pre-processing Methods 0.000 claims abstract description 4
- 238000013507 mapping Methods 0.000 claims description 12
- 238000006243 chemical reaction Methods 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 3
- 230000004927 fusion Effects 0.000 abstract description 7
- 230000000694 effects Effects 0.000 abstract description 3
- 238000003384 imaging method Methods 0.000 description 13
- 230000003287 optical effect Effects 0.000 description 7
- 238000000034 method Methods 0.000 description 6
- 238000009434 installation Methods 0.000 description 3
- 239000003550 marker Substances 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 241000251468 Actinopterygii Species 0.000 description 1
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 210000000746 body region Anatomy 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000005670 electromagnetic radiation Effects 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000002844 melting Methods 0.000 description 1
- 230000008018 melting Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration by the use of more than one image, e.g. averaging, subtraction
-
- G06T5/80—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
- G06T2207/10036—Multispectral image; Hyperspectral image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Abstract
A kind of fusion method of terahertz image and visible images, belongs to Multispectral Image Fusion field, it is characterised in that:By terahertz image acquisition device and visible light image acquisition device at target center be overlapped setting;Detected body is shot respectively, and the terahertz image and visible images for obtaining detected body carry out image preprocessing;Pretreated terahertz image and visible images are subjected to image registration operation, extract characteristics of image, and geometric correction is carried out to visible images;It is merged according to the result of registration operation, obtains completely detected body target area.Detected body is shot by terahertz image acquisition device and visible light image acquisition device and carries out image information collecting, so that the information of image is richer after merging, details is more comprehensively.The effect of image co-registration is fine, combines the information of visible images and terahertz image, realize to object scene more comprehensively, clearly indicate, can be applied to the various fields such as security protection.
Description
Technical field
The invention belongs to the fusions of Multispectral Image Fusion field more particularly to a kind of terahertz image and visible images
Method.
Background technique
Terahertz image and visible images merge the scope for belonging to Multispectral Image Fusion, the electromagnetic wave of different-waveband
It is formed by that feature of image is different by reflecting or transmiting, the target information provided is different, using the complementation of this information,
Realize the display more apparent to each state of target.Terahertz (Terahertz) radiation refers to frequency of oscillation in 1012Hz (1THz
=1012GHz) left and right wave band electromagnetic radiation.The electromagnetic wave of this frequency range just between millimeter wave and it is infrared between, therefore light
Spectrum field is known as far ir ray (FIR-far infrared);It is the millimeter wave (MMW) or submillimeter wave of person in electronics
(SMMW).In general, the frequency range of terahertz wave band is defined as 0.1THz-10THz (wavelength:3mm-30μm).THz wave
With good penetrability, the cache in clothes can be detected;The substances such as safety check drugs of interest, explosive are in Terahertz
Wave band has a characteristic spectrum, and passive type terahertz imaging not actively outside radiated electromagnetic wave, will not cause electricity to tested personnel
From injury, it is suitble to human body safety check imaging.Since THz wave is longer than visible light and infrared light wavelength, the Terahertz caused
Fogging image.
Visible images have objective contour details abundant and color characteristic.It can be mentioned in the environment of illumination good brightness
For the complexion of target object, profile and color information, visual experience comfortable to human eye while, provide target details abundant
Information.
Since Terahertz is different from the imaging circumstances of visible images, the installation situation of image collecting device is different, imaging
Optical path is different, and Terahertz and the resolution ratio and Pixel size of visual light imaging plane are different, and institute is at Terahertz and visible light figure
Situations such as picture will appear imaging region difference, and imaging plane central point is not overlapped.These problems can all lead to not directly carry out
Information fusion between image object profile, simultaneously because the difference and limitation of Terahertz and visible images, rely solely on list
The image of one type, it is difficult to meet the actual demand in engineering.
Summary of the invention
The present invention is solving the above problems, and provides the fusion method of a kind of Terahertz and visible images.
The fusion method of terahertz image of the present invention and visible images, including:
By terahertz image acquisition device and visible light image acquisition device at target center be overlapped setting;
Detected body is shot by terahertz image acquisition device and visible light image acquisition device respectively, is detected
The terahertz image and visible images of body;
Image preprocessing is carried out to above-mentioned terahertz image and visible images respectively;
Pretreated terahertz image and visible images are subjected to image registration operation, extract characteristics of image, and right
Visible images carry out geometric correction;
According to registration operation as a result, being merged to terahertz image and visible light image information, in terahertz image
Middle acquisition is completely detected body target area.
The fusion method of terahertz image of the present invention and visible images, the registration operation are by terahertz
The hereby mapping of image and visible images on space and gray scale;The human region that terahertz image and visible images will acquire
Image normalization is same size, using visible images human region, to human region incomplete in terahertz image into
Row supplement is complete, to obtain complete human body terahertz image.
The concrete mode of the mapping is:Terahertz image and visible images are expressed as a two-dimensional sequence I1
(x, y) and I2(x, y);Wherein I1(x, y) and I2(x, y) respectively indicates the gray value on corresponding position (x, y);
Mapping between terahertz image and visible images is represented by:
I2(x, y)=g (I1(f (x, y)))
Wherein f be terahertz image and visible images coordinate system coordinate conversion relation function, be expressed as (x ', y ')=
f(x,y);G is the grey scale mapping relationship of terahertz image and visible images.
The fusion method of terahertz image of the present invention and visible images, the geometric correction of the visible images
Concrete mode be:The marginal information characteristic point of visible images and the characteristic point of label are obtained, using polynomial mode meter
Distortion parameter is calculated, characteristic point is substituted into binary polynomial and solves distortion parameter;
The distortion parameter of visible images is calculated using binary quadratic polynomial, image coordinate point (x, y) and distortion
Relationship between image coordinate (x ', y '):
X '=α00+α10x+α01y+α20x2+α11xy+α02y2
Y '=b00+b10x+b01y+b20x2+b11xy+b02y2
In formula:Parameter x is the abscissa of original image, and parameter y is the ordinate of original image, and parameter x ' is the cross of fault image
Coordinate, parameter y ' are the ordinates of fault image.α00, α10, α01, α20, α11, α02It is the distortion parameter of image abscissa, b00,
b10, b01, b20, b11, b02It is the distortion parameter of image ordinate.
According to above-mentioned formula, the distortion parameter of visible images is found out, distortion correction is carried out to image.
The fusion method of terahertz image of the present invention and visible images, described image pretreatment mode include going
It makes an uproar, be filtered and binary conversion treatment.
The fusion method of terahertz image of the present invention and visible images is acquired by terahertz image fill respectively
It sets and shoots detected body progress image information collecting with visible light image acquisition device, fill two kinds after image procossing is carried out to it
It sets acquired image information to be merged, the fusion method of terahertz image of the present invention and visible images, so that melting
The information of image is richer after closing, and details is more comprehensively.The effect of image co-registration is fine, combines visible images and Terahertz
The information of image, realize to object scene more comprehensively, clearly indicate, can be applied to the various fields such as security protection.
Detailed description of the invention
Fig. 1 is the fusion method schematic diagram of terahertz image of the present invention and visible images;
Fig. 2 is image collecting device coordinate system schematic illustration of the present invention;
Fig. 3 is twiddle factor coordinate schematic illustration of the present invention.
Specific embodiment
With reference to the accompanying drawings and embodiments, the fusion method for inventing the terahertz image and visible images is carried out detailed
Describe in detail bright, the detected body of following embodiments is human body.
Embodiment 1
As shown in Figure 1, the fusion method of terahertz image of the present invention and visible images, includes the following steps:
1) by terahertz image acquisition device and visible light image acquisition device be overlapped setting at the center of target;Benefit
With the image information of Terahertz, guarantee the integrality of terahertz image information, the validity of image co-registration, and obtains completely quilt
Detect body region;
2) detected body is shot by terahertz image acquisition device and visible light image acquisition device respectively, acquisition is visited
Survey the terahertz image and visible images of body;
3) image preprocessing is carried out to above-mentioned terahertz image and visible images respectively;Described image pretreatment mode packet
Include denoising, filtering processing and binary conversion treatment;Since the passive type mode that terahertz image uses obtains image, image is by ring
The influence of the factors such as border, temperature can include a large amount of noise in image;Image is to be scanned imaging using radiometer simultaneously,
Partial region can lose some pixels, and the information for causing image to obtain is imperfect.
Smooth, progress image enhancement is carried out to terahertz image by the way of filtering, so that the profile of target is apparent,
Then edge detection is carried out, human body target region is obtained, coarse positioning is carried out to human region, human region is then based on and carries out people
The fine positioning of body.
Visible images carry out smothing filtering and remove noise, and the area of human body is obtained by marginal information and human body feature point
Domain.
4) pretreated terahertz image and visible images are subjected to image registration operation, extract characteristics of image, and
Geometric correction is carried out to visible images;
5) according to registration operation as a result, being merged to terahertz image and visible light image information, in Terahertz figure
Completely detected body target area is obtained as in.
The present embodiment obtains terahertz image by terahertz image acquisition device and visible light image acquisition device and can
Light-exposed image clicks through line distortion correction using marker characteristic, by the visible images and terahertz image progress target after correction
Fusion obtains the human body contour outline information of closely complete terahertz image.The Terahertz equipment that the present embodiment uses is greater than
The THz wave that the radiometer in 16 channels generates human body absorbs, and is then imaged;Visible light camera uses wide-angle or fish
Eye imaging head obtains the visible images between 1-2 meters.
Embodiment 2
On the basis of example 1, different with the installation situation of visible light image acquisition device due to Terahertz, two pass
The optical path that sensor passes through is not identical and field range of camera lens is different, can go out between terahertz image and visible images
Now opposite translation, rotation, imaging region be not of uniform size, the different problem of resolution sizes, so that between two images
It cannot directly be merged, it is necessary to carry out image registration processing, two images are transformed to below the same coordinate system.Such as Fig. 2 institute
Show, X in figurec, Yc, ZcImage collecting device coordinate system is formed, C point is image collecting device optical center, ZcAxis is image collecting device
Optical axis, the plane vertical with optical axis be image collecting device imaging plane, x, y be imaging plane coordinate system, optical axis at
As the intersection point O of plane|The as central point of image.OO|For image collecting device focal length.If Terahertz is adopted with visible images
Acquisition means can make the central point of imaging region not be overlapped because installation situation difference causes optical axis that can not be completely coincident, two width figures
As having differences under plane coordinate system, need that they are registrated under same plane coordinate system by corresponding geometric transformation
Face.
The registration operation is the mapping by terahertz image and visible images on space and gray scale;
The concrete mode of the mapping is:Terahertz image and visible images are expressed as a two-dimensional sequence I1
(x, y) and I2(x, y);Wherein I1(x, y) and I2(x, y) respectively indicates the gray value on corresponding position (x, y);
Mapping between terahertz image and visible images is represented by:
I2(x, y)=g (I1(f (x, y)))
Wherein f be terahertz image and visible images coordinate system coordinate conversion relation function, be expressed as (x ', y ')=
f(x,y);G is the grey scale mapping relationship of terahertz image and visible images.
As shown in figure 3, giving two coordinate systems of a rotation angle, it is assumed that the rotation angle of difference is Δ θ, P point
Coordinate representation under terahertz image coordinate system is (R х cos θ, R х sin θ), then the coordinates table under visible light coordinate system
It is shown as [R х cos (θ-Δ θ), R х sin (θ-Δ θ)].Wherein R is distance of the coordinate origin to P point, terahertz image and visible
The rotation differential seat angle Δ θ of light is within ± 10 degree.
In input picture f (x, y), gray value is only defined at integer position (x, y), by obtaining at spatial alternation
The gray value of new images g (x, y) determine that the present apparatus uses bilinearity by the value for the f (x, y) being on non-integer position
Interpolation obtains new image.
Concrete mode for the geometric correction of visible images is:Obtain visible images marginal information characteristic point and
The characteristic point of label calculates distortion parameter using polynomial mode, substitutes into characteristic point to binary polynomial and solves distortion parameter;
The distortion parameter of visible images is calculated using binary quadratic polynomial, image coordinate point (x, y) and distortion
Relationship between image coordinate (x ', y '):
X '=α00+α10x+α01y+α20x2+α11xy+α02y2
Y '=b00+b10x+b01y+b20x2+b11xy+b02y2
According to above-mentioned formula, the distortion parameter of visible images is found out, distortion correction is carried out to image.Pass through the above coordinate
The mismatch problem of Terahertz and visible images can be converted into comprising scale factor, translational movement, rotation by the conversion between system
The mathematical formulae of the factor.Terahertz and visible images two width from different coordinates, by the conversion for calculating the two
Parameter is registrated in the same coordinate system.Marginal information characteristic point is obtained using sift feature extraction, and the characteristic point of label is using black
Angle point on the image zooming-out visible images at white interval calculates the point of physical location as characteristic point, brings formula calculating into and obtains
Take the parameter of correction image.Using the parameter, the visible images of visible light camera acquisition are calculated.
Terahertz image of the present invention and the fusion method of visible images are by terahertz image and visible images
The coaxial setting of focus, it is seen that light image carries out polynomial computation using marker characteristic point, distortion parameter is obtained, to visible images
Distortion correction is carried out, by being registrated to terahertz image and visible images, obtains complete human body in terahertz image
Profile information.
Conventional point feature registration method be based on grey scale change feature is described, there are the big Terahertz of gray difference with
Visible light image registration failure, terahertz image syncretizing effect and fusion accuracy are dry vulnerable to terahertz image image quality and noise
The problem of disturbing influence, the method for the invention use the method for registering based on Edge Feature Points and marker characteristic point, pass through label
Characteristic point and marginal information characteristic point are corrected image.Reduce or eliminate the relative displacement between image, rotation and scale
The geometric deformations such as transformation, to obtain the image with Space Consistency.Combining Multi-scale corner detection and the main side of characteristic point
It is improved to definition, increases rotational invariance and scale invariability, while guaranteeing registration accuracy and method stability
Improve efficiency.Detected body extraction algorithm is improved in conjunction with the Edge texture information feature of visible light, improves image
Quality so that Terahertz edge contour is more complete.The detected body target area for accurately obtaining terahertz image is subsequent
Human body target dangerous material image detection lays the foundation.
Claims (4)
1. the fusion method of a kind of terahertz image and visible images, it is characterised in that including:
By terahertz image acquisition device and visible light image acquisition device at target center be overlapped setting;
Detected body is shot by terahertz image acquisition device and visible light image acquisition device respectively, obtains detected body
Terahertz image and visible images;
Image preprocessing is carried out to above-mentioned terahertz image and visible images respectively;
Pretreated terahertz image and visible images are subjected to image registration operation, extract characteristics of image, and to visible
Light image carries out geometric correction;
According to registration operation as a result, merged to terahertz image and visible light image information, obtained in terahertz image
Take whole detected body target area.
2. the fusion method of terahertz image according to claim 1 and visible images, it is characterised in that:Described matches
Quasi- operation is the mapping by terahertz image and visible images on space and gray scale;
The concrete mode of the mapping is:Terahertz image and visible images are expressed as a two-dimensional sequence I1(x, y)
And I2(x, y);Wherein I1(x, y) and I2(x, y) respectively indicates the gray value on corresponding position (x, y);
Mapping between terahertz image and visible images is represented by:
I2(x, y)=g (I1(f (x, y)))
Wherein f be terahertz image and visible images coordinate system coordinate conversion relation function, be expressed as (x ', y ')=f (x,
y);G is the grey scale mapping relationship of terahertz image and visible images.
3. the fusion method of terahertz image according to claim 1 or 2 and visible images, it is characterised in that:It is described
The concrete mode of the geometric correction of visible images is:Obtain the marginal information characteristic point of visible images and the feature of label
Point calculates distortion parameter using polynomial mode, substitutes into characteristic point to binary polynomial and solves distortion parameter;
The distortion parameter of visible images is calculated using binary quadratic polynomial, image coordinate point (x, y) and fault image
Relationship between coordinate (x ', y '):
X '=a00+a10x+α01y+α20x2+a11xy+a02y2
Y '=b00+b10x+b01y+b20x2+b11xy+b02y2
According to above-mentioned formula, the distortion parameter of visible images is found out, distortion correction is carried out to image.
4. the fusion method of terahertz image according to claim 3 and visible images, it is characterised in that:Described image
Pretreatment mode includes denoising, filtering processing and binary conversion treatment.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810654322.6A CN108846823A (en) | 2018-06-22 | 2018-06-22 | A kind of fusion method of terahertz image and visible images |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810654322.6A CN108846823A (en) | 2018-06-22 | 2018-06-22 | A kind of fusion method of terahertz image and visible images |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108846823A true CN108846823A (en) | 2018-11-20 |
Family
ID=64203471
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810654322.6A Pending CN108846823A (en) | 2018-06-22 | 2018-06-22 | A kind of fusion method of terahertz image and visible images |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108846823A (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109492714A (en) * | 2018-12-29 | 2019-03-19 | 同方威视技术股份有限公司 | Image processing apparatus and its method |
CN110414470A (en) * | 2019-08-05 | 2019-11-05 | 深圳市矽赫科技有限公司 | Visiting method based on Terahertz and visible light |
CN112346141A (en) * | 2020-11-05 | 2021-02-09 | 上海亨临光电科技有限公司 | Terahertz image and visible light image mapping fusion method and system |
CN112381748A (en) * | 2020-11-16 | 2021-02-19 | 上海亨临光电科技有限公司 | Terahertz and visible light image registration method and device based on texture feature points |
CN112381747A (en) * | 2020-11-16 | 2021-02-19 | 上海亨临光电科技有限公司 | Terahertz and visible light image registration method and device based on contour feature points |
WO2021031626A1 (en) * | 2019-08-19 | 2021-02-25 | 同方威视技术股份有限公司 | Method and device for image processing, computer system and readable storage medium |
CN112508113A (en) * | 2020-12-14 | 2021-03-16 | 中国科学院空天信息创新研究院 | Method and device for detecting passive terahertz human body image hidden target |
CN113126176A (en) * | 2019-12-30 | 2021-07-16 | 清华大学 | Terahertz wave security inspection system and method |
CN113538405A (en) * | 2021-07-30 | 2021-10-22 | 吉林大学 | Nondestructive testing method and system for glass fiber composite material based on image fusion |
CN113807286A (en) * | 2021-09-24 | 2021-12-17 | 福建平潭瑞谦智能科技有限公司 | Face recognition big data training method |
CN116818704A (en) * | 2023-03-09 | 2023-09-29 | 苏州荣视软件技术有限公司 | High-precision full-automatic detection method, equipment and medium for semiconductor flaw AI |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106124435A (en) * | 2016-07-04 | 2016-11-16 | 江苏大学 | Based on visible ray, near-infrared, the rice new-old quality inspection device of Terahertz fusion spectral technique and detection method |
-
2018
- 2018-06-22 CN CN201810654322.6A patent/CN108846823A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106124435A (en) * | 2016-07-04 | 2016-11-16 | 江苏大学 | Based on visible ray, near-infrared, the rice new-old quality inspection device of Terahertz fusion spectral technique and detection method |
Non-Patent Citations (2)
Title |
---|
乔玉龙: ""太赫兹与可见光双波段图像的配准与融合研究"", 《中国硕士学位论文全文数据库》 * |
曹玲玲: ""基于多项式的遥感图像快速几何校正"", 《电脑开发与应用》 * |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109492714B (en) * | 2018-12-29 | 2023-09-15 | 同方威视技术股份有限公司 | Image processing apparatus and method thereof |
CN109492714A (en) * | 2018-12-29 | 2019-03-19 | 同方威视技术股份有限公司 | Image processing apparatus and its method |
CN110414470A (en) * | 2019-08-05 | 2019-11-05 | 深圳市矽赫科技有限公司 | Visiting method based on Terahertz and visible light |
WO2021031626A1 (en) * | 2019-08-19 | 2021-02-25 | 同方威视技术股份有限公司 | Method and device for image processing, computer system and readable storage medium |
CN113126176A (en) * | 2019-12-30 | 2021-07-16 | 清华大学 | Terahertz wave security inspection system and method |
CN112346141A (en) * | 2020-11-05 | 2021-02-09 | 上海亨临光电科技有限公司 | Terahertz image and visible light image mapping fusion method and system |
CN112381748A (en) * | 2020-11-16 | 2021-02-19 | 上海亨临光电科技有限公司 | Terahertz and visible light image registration method and device based on texture feature points |
CN112381747A (en) * | 2020-11-16 | 2021-02-19 | 上海亨临光电科技有限公司 | Terahertz and visible light image registration method and device based on contour feature points |
CN112508113A (en) * | 2020-12-14 | 2021-03-16 | 中国科学院空天信息创新研究院 | Method and device for detecting passive terahertz human body image hidden target |
CN113538405B (en) * | 2021-07-30 | 2023-03-31 | 吉林大学 | Nondestructive testing method and system for glass fiber composite material based on image fusion |
CN113538405A (en) * | 2021-07-30 | 2021-10-22 | 吉林大学 | Nondestructive testing method and system for glass fiber composite material based on image fusion |
CN113807286A (en) * | 2021-09-24 | 2021-12-17 | 福建平潭瑞谦智能科技有限公司 | Face recognition big data training method |
CN116818704A (en) * | 2023-03-09 | 2023-09-29 | 苏州荣视软件技术有限公司 | High-precision full-automatic detection method, equipment and medium for semiconductor flaw AI |
CN116818704B (en) * | 2023-03-09 | 2024-02-02 | 苏州荣视软件技术有限公司 | High-precision full-automatic detection method, equipment and medium for semiconductor flaw AI |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108846823A (en) | A kind of fusion method of terahertz image and visible images | |
Ayhan et al. | A novel utilization of image registration techniques to process mastcam images in mars rover with applications to image fusion, pixel clustering, and anomaly detection | |
US11838490B2 (en) | Multimodal imaging sensor calibration method for accurate image fusion | |
WO2021071995A1 (en) | Systems and methods for surface normals sensing with polarization | |
Steiner et al. | Design of an active multispectral SWIR camera system for skin detection and face verification | |
CN108731587A (en) | A kind of the unmanned plane dynamic target tracking and localization method of view-based access control model | |
Teke | Satellite image processing workflow for RASAT and Göktürk-2 | |
CN109308702A (en) | A kind of real-time recognition positioning method of target | |
CN104318550A (en) | Eight-channel multi-spectral imaging data processing method | |
CN110210292A (en) | A kind of target identification method based on deep learning | |
CN107356338A (en) | A kind of the infrared remote sensing imaging device and method of big breadth high temperature sensitivity | |
Wei et al. | RegiNet: Gradient guided multispectral image registration using convolutional neural networks | |
Weinmann et al. | Fusing passive and active sensed images to gain infrared-textured 3D models | |
Jabari et al. | Improving UAV imaging quality by optical sensor fusion: an initial study | |
US11568683B2 (en) | Facial biometrics system and method using digital fingerprints | |
Liu et al. | An improved APAP algorithm via line segment correction for UAV multispectral image stitching | |
Zhao et al. | Assessment of SPOT-6 optical remote sensing data against GF-1 using NNDiffuse image fusion algorithm | |
Karaca et al. | Ground-based panoramic stereo hyperspectral imaging system with multiband stereo matching | |
Gu et al. | Wireless sensor system of UAV infrared image and visible light image registration fusion | |
Zou et al. | Research on seamless image stitching based on depth map | |
Hakim et al. | Satellite Attitude Determination Based on Pushbroom Image Band Coregistration | |
Liu et al. | Multitask saliency detection model for synthetic aperture radar (SAR) image and its application in SAR and optical image fusion | |
Cronje et al. | A comparison of image features for registering LWIR and visual images | |
Yu et al. | An infared polarization image fusion method based on NSCT and fuzzy C-means clustering segmentation algorithms | |
Rogelj et al. | Reducing object curvature and height variation effects in hyperspectral images |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181120 |